• Journal of the Mineralogical Society of Korea
• /
• v.17 no.4
• /
• pp.365-383
• /
• 2004

Granitic rocks in the southeastern Gyeongsang Basin can be classified into three groups. The group I contains various mafic microgranular enclave (MME) and/or mafic clot which implies magma mixing or mingling. The group II show the feature of shallow depth emplacement at low pressure, and the group III is characterized by A-type granite implying extensional tectonic environment. Mineralogical characteristics of the granitic rocks have showed systematic variations in perthite exsolution temperatures and biotite compositions according to their rock facies, although they do not show any distinctively different trend in geography and textures or rock facies. Amphiboles from Group I are calcic-amphibole and they were formed at 0.4 ～ 2.8 kb in pressure based on the amphibole geobarometry. Amphiboles from group ill are riebeckite, whileas amphiboles were not observed in Group II. The chemical composition of biotite defined in clusters showing a continuous spectrum between group I to ferric-annite of group ill. The composition of plagioclase generally plotted in albite, oligoclase, and andesine area without any distinctive differences among their geography or rock facies. The exsolution temperatures by perthite geothermometry are calculated as $300～400^{\circ}C$ in Group I, and 500～$600^{\circ}C$ in equigranular granite of group II and alkali-feldspar granite of group III.

• Journal of the Mineralogical Society of Korea
• /
• v.17 no.1
• /
• pp.37-47
• /
• 2004

Sulfates such as alunite and schwertmannite formed under oxidation condition play a important role in geochemical processes taken place at waste dumps and a creek from Dalseong mine, Daegu. Water chemistry shows pH decreases from upstream toward downstream creek, mainly due to formation of schwertmannite that is the most abundant phase along the creek. The removal of Al from the creek is preferentially attributed to formation of Al-bearing minerals and Al-sulphates. Among them, alunite is the most important Al-sink phase that occurs at higher pH than $pK_1$, Al hydrolysis constant. With high saturation index, alunite formed at the creek has a spherical form, commonly associated with schwertmannite. Secondary minerals formed on the surface of altered or weathered surfaces of heavy metals from the wasted dump that underwent severe oxidation, where alunite has characteristic habits which are spheric, radiating, and botrytis-like aggregates. Natroalunite occurs in association with alunite, or as mixtures of both of them. Because the pH decreases with distance due to formation of schwertmannite, although total contents of dissolved ions slowly lessen at least in the AMD, it is expected that the minerals precipitated at the creek can be exposed to subsequent dissolution, which may induce possible environmental problems.

• Journal of KIISE:Software and Applications
• /
• v.34 no.9
• /
• pp.857-868
• /
• 2007

The good classifier ensemble should have a high complementarity among classifiers in order to produce a high recognition rate and its size is small in order to be efficient. This paper proposes a classifier ensemble selection algorithm with coarse-to-fine stages. for the algorithm to be successful, the original classifier pool should be sufficiently diverse. This paper produces a large classifier pool by combining several different classification algorithms and lots of feature subsets. The aim of the coarse selection is to reduce the size of classifier pool with little sacrifice of recognition performance. The fine selection finds near-optimal ensemble using genetic algorithms. A hybrid genetic algorithm with improved searching capability is also proposed. The experimentation uses the worldwide handwritten numeral databases. The results showed that the proposed algorithm is superior to the conventional ones.

• Journal of KIISE:Software and Applications
• /
• v.37 no.2
• /
• pp.85-94
• /
• 2010

This paper introduces a novel way to leverage convolution parse tree kernel to extract the interaction information between two proteins in a sentence without multiple features, clues and complicated kernels. Our approach needs only the parse tree alone of a candidate sentence including pairs of protein names which is potential to have interaction information. The main contribution of this paper is two folds. First, we show that for the PPI, it is imperative to execute parse tree pruning removing unnecessary context information in deciding whether the current sentence imposes interaction information between proteins by comparing with the latest existing approaches' performance. Secondly, this paper presents that tree kernel decay factor can play an pivotal role in improving the extraction performance with the identical learning conditions. Consequently, we could witness that it is not always the case that multiple kernels with multiple parsers perform better than each kernels alone for PPI extraction, which has been argued in the previous research by presenting our out-performed experimental results compared to the two existing methods by 19.8% and 14% respectively.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.38 no.1
• /
• pp.107-118
• /
• 2010

Palaces of a country can be representative aggregate of the culture and arts of the country. Palaces were used not only as a living place to protect kings and royal families and to live cultural lives, but also a political place to govern the country. Kings in Choseon founded the country on the Sung Confucianism, their philosophical background. They built Bangji and Wondo as constant reminders of their philosophy. Bangji is the most apparent characteristic in the palace gardens of the Choseon Dynasty. Finding out the original form of Jidang, restoring the original and passing it on to future generations can be passing on the spiritual culture of our ancestors. This study is done to find out or locate well preserved Jidang, transformed Jidang and lost Jidang in Changdeokgung, which is the best conserved palace of Choseon. This study is composed of Jidang which has been kept in its original state, transformed Jidang, locating lost Jidang and Jidnang which has been unearthed. The total number of Donggwoldo's Jidangs is 17; those that are kept in their original state, 4; transformed Jidangs, 5; Jidangs that were lost and have been located, 8; and newly appeared Jidangs, 2. From the results, opinions on restoration are presented. This study can be a small drop in the thin stream of tradition passing onto future generations in this world where knowledge and information are transported momentarily and the classification of culture and border is mixed, yielding vagueness.

• The KIPS Transactions:PartD
• /
• v.13D no.7 s.110
• /
• pp.985-996
• /
• 2006

As the internet technologies develop, the geographic information system environment is changing to the web-based service. Since geospatial information of the existing Web-GIS services were developed independently, there is no interoperability to support diverse map formats. In spite of the same geospatial information object it can be used for various proposes that is duplicated in GIS separately. It needs intelligent strategies for optimal replica selection, which is identification of replication geospatial information objects. And for management of replication objects, OMG, GLOBE and GRID computing suggested related frameworks. But these researches are not thorough going enough in case of geospatial information object. This paper presents a model of location service, which is supported for optimal selection among replication and management of replication objects. It is consist of tree main services. The first is binding service which can save names and properties of object defined by users according to service offers and enable clients to search them on the service of offers. The second is location service which can manage location information with contact records. And obtains performance information by the Load Sharing Facility on system independently with contact address. The third is intelligent selection service which can obtain basic/performance information from the binding service/location service and provide both faster access and better performance characteristics by rules as intelligent model based on rough sets. For the validity of location service model, this research presents the processes of location service execution with Graphic User Interface.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.37 no.4
• /
• pp.225-232
• /
• 2024

This study focuses on the determination of the fracture mode in composite laminates using deep learning. With the increase in the use of laminated composites in numerous engineering applications, the insurance of their integrity and performance is of paramount importance. However, owing to the complex nature of these materials, the identification of fracture modes is often a tedious and time-consuming task that requires critical domain knowledge. Therefore, to alleviate these issues, this study aims to utilize modern artificial intelligence technology to automate the fractographic analysis of laminated composites. To accomplish this goal, scanning electron microscopy (SEM) images of fractured tensile test specimens are obtained from laminated composites to showcase various fracture modes. These SEM images are then categorized based on numerous fracture modes, including fiber breakage, fiber pull-out, mix-mode fracture, matrix brittle fracture, and matrix ductile fracture. Next, the collective data for all classes are divided into train, test, and validation datasets. Two state-of-the-art, deep learning-based pre-trained models, namely, DenseNet and GoogleNet, are trained to learn the discriminative features for each fracture mode. The DenseNet models shows training and testing accuracies of 94.01% and 75.49%, respectively, whereas those of the GoogleNet model are 84.55% and 54.48%, respectively. The trained deep learning models are then validated on unseen validation datasets. This validation demonstrates that the DenseNet model, owing to its deeper architecture, can extract high-quality features, resulting in 84.44% validation accuracy. This value is 36.84% higher than that of the GoogleNet model. Hence, these results affirm that the DenseNet model is effective in performing fractographic analyses of laminated composites by predicting fracture modes with high precision.